Leading a paediatric resuscitation team — OSCE

Station brief (8–10 minutes)

A four-year-old arrests on the paediatric ward at three in the morning. You arrive first as the senior registrar. Two nurses, a resident and a pharmacist are at the bed space, and the parents are watching from the doorway. The examiner will ask you to take the team leader role and describe, step by step, how you would run this resuscitation — and then how you would build a team that performs this well when you are not standing in it. You will be questioned on role allocation, closed-loop communication, the shared mental model, psychological safety and speak-up, the TeamSTEPPS toolkit, structured handover, and the evidence for team training. Do not invent local staffing ratios or mandated policy numbers. [4] [11]

Tasks for the candidate

1. Take the team leader role and describe your first 60 seconds. [4]
2. Demonstrate closed-loop communication for a critical drug dose and explain why it matters. [2]
3. Describe how you would build and maintain a shared mental model and flatten the authority gradient. [4] [11]
4. State how you would hand the child over to PICU and how you would sustain the team after the event. [13]

Expected performance

Must hit. Declares the team leader role out loud; allocates roles by name and task (airway, compressions, drugs, scribe); uses closed-loop communication (message, read-back, confirm) for every dose; describes a structured pause each minute to share the mental model (who the child is, working diagnosis, plan, roles, next contingency); names psychological safety as the enabler of voice and states at least one concrete action to flatten the authority gradient (invite speak-up by name and respond well the first time); hands over with I-PASS and names the receiver's synthesis as the closed loop; commits to a structured debrief and to supporting the second victim. [2] [4] [13]

Merit. Names the TeamSTEPPS toolkit (SBAR, CUS, two-challenge rule, check-back, huddle, brief, debrief) and says when each is used; distinguishes leadership from management and matches an authoritative style in crisis to a facilitative style in complex longitudinal care; cites the Starmer I-PASS evidence and the Salas team-training meta-analysis; cites the Cheng paediatric simulation-based CRM framework and the Manser dynamic-domains review; cites the Cochrane interprofessional collaboration review with its honest caveat about heterogeneous studies; names a care coordinator and a written shared plan for a medically complex child. [3] [5] [11] [14]

Fail. Fails to name a leader or to allocate roles; uses open-loop orders with no read-back; has no shared mental model; cannot name a tool to flatten the authority gradient or respond to a dismissed concern; hands over with no structure; proposes no debrief; confuses leadership with management or interprofessional collaboration with holding more meetings; or invents local staffing ratios or mandated thresholds. [2] [4] [13]

Sample candidate structure

"I am taking the role of team leader. Sarah, you are airway; David, you are doing compressions; Maria, you are drugs — draw up each dose as I call it and read it back to me; Tom, you are scribe and time-keeper. Can someone call the PICU consultant and ask a nurse to stay with the parents and keep them informed? I am standing back so I can see the whole child and the monitors and think ahead — I will not be doing compressions, because a leader doing a task has stopped leading. For every dose I will use closed-loop communication: I call 'adrenaline, ten micrograms', Maria reads back 'adrenaline ten micrograms, drawing up now', and I confirm before it is given — that loop is what stops a dose being missed or doubled. Every minute or two I will pause the team to share the mental model: this is a four-year-old, pulseless, the rhythm is shockable, we have shocked twice, the plan is a third shock then adrenaline, Maria is on drugs, the next contingency is to consider reversible causes. A shared mental model keeps us coordinated as the picture changes. To flatten the authority gradient I will say out loud 'if anyone sees something wrong, please say it — that includes the student' — and the first time someone does, I will thank them and act on it, because a poor response silences the team forever. If a concern is voiced and dismissed once, the two-challenge rule means we voice it again to force a hearing. When we hand over to PICU I will use I-PASS — illness severity, patient summary, action list, situation awareness and contingency planning, and synthesis by the receiver — and the receiver reads the plan back, which is the closed loop of handover. After the event I will run a short structured debrief — what went well, what we learned, what we change — and I will check on the team and the parents, because a clinician harmed by an adverse event is a second victim who needs support, not blame. To build a team that performs like this when I am not here, I embed TeamSTEPPS in daily work, standardise handover with I-PASS and audit it, and reinforce the skills with simulation-based crew resource management and debrief, because the skills decay without reinforcement. The evidence supports this: the Salas meta-analysis shows team training improves team performance, Starmer showed I-PASS reduced medical errors, and the Cochrane review on interprofessional collaboration shows these interventions can improve outcomes — though I read that evidence honestly, because the studies are heterogeneous and the effect size is context-dependent." [2] [3] [4] [13]